Haute Lumière

Commerce · IV.01 · MMXXVI · daylight

La Bourse  /  Volume IV  /  Nº IV.01  /  Workbook — the student

WORKBOOK — THE STUDENT

Chapter IV.01 · Making Things That Add

For the person learning this for themselves. A term of practice, a term project, and a way of checking your own progress. Applied to a life, and to the objects in it.

---

WHY THIS WORKBOOK IS DIFFERENT

Most environmental education asks you to feel something about consumption. This one asks you to read a document.

The skill in this chapter is unusually portable and unusually rare: you learn to take a published environmental claim, find the four declarations behind it — boundary, baseline, counterfactual, unit — and recompute the claim yourself from the maker's own numbers. It takes about ninety minutes to learn and you keep it for life. It works on carpet, on cars, on food, on electricity, on a university's own net-zero pledge, and on your own habits.

You need no laboratory, no subscription and no permission. Environmental product declarations are public documents, published deliberately, by manufacturers who paid to have them verified. The whole exercise is reading what somebody already wrote down and hoped you would find useful.

One rule before you start. You are not hunting for liars. In every case in this chapter the manufacturer published the number that complicates their own headline. The gap is between the document and the summary of the document, and the summary is what travels. Approach it as a reader, not a prosecutor, and you will find much more.

---

PART ONE — DISCOVERY

Weeks 1–4: find the declarations that already exist

Exercise 1.1 — Your first declaration (90 minutes)

Choose one manufactured object you can see from where you are sitting. Search for <manufacturer> <product> environmental product declaration or <product> EPD. Download the PDF. Then, before reading anything else, find and write down four things:

1. The functional unit, and the units it is in.
2. The reference service life, in years.
3. Which modules are declared — and, importantly, which are not.
4. Whether any use-stage module is declared for a shorter period than the service life.

If you cannot find all four, that is a finding and you should write it down as one. A declaration that omits its service life is telling you something.

Exercise 1.2 — The annualisation (30 minutes)

Take the use-stage module and multiply it by the service life, the way the chapter does:

  B2 x reference service life  =  whole-life maintenance
  0.403 x 15                   =  6.045 kg CO2e/m2

Then recompute the whole-life total and the module shares. Write the two percentage tables side by side, as the chapter does, and note which module was largest before you multiplied and which is largest after.

For the worked product the answer moves from A1–A3 at 72.0 percent to B2 at 63.8 percent. Yours may not move at all. Both outcomes are worth writing down; a declaration whose shares do not move is a well-constructed one and you should note the manufacturer's name.

Exercise 1.3 — The appreciative sweep (one week, background)

Over a week, collect five claims you meet in ordinary life — a label, an advertisement, a university pledge, a delivery service's footer, a friend's confident sentence. For each, write one line naming which of the four declarations is present and which is missing.

Do not argue with anyone. You are calibrating, not debating. By the end of the week you will have noticed something that surprises most people: the most carefully-qualified claims usually come from the organisations doing the most work, and the vaguest come from the ones doing the least. Vagueness is the tell, not exaggeration.

---

PART TWO — THE ARITHMETIC

Weeks 5–8: compute before you have an opinion

Exercise 2.1 — The six boundaries (2 hours)

Rebuild the chapter's central table for your own product. Six rows, each a defensible boundary, each with its own number:

  gate stage, with any biogenic credit taken
  gate stage, as declared
  gate stage, biogenic credit refused
  the declared boundary, as printed
  the declared boundary, service life corrected
  the same, with the least favourable end-of-life scenario

Then compute the spread and note whether the sign changes anywhere. For the chapter's tile the spread is 11.625 kg CO₂e/m² across a single square metre. Whatever yours is, write the sentence underneath: the physics did not move; the line moved.

Exercise 2.2 — Implied abatement cost (one hour)

Find two versions of the same product at different prices with different declared footprints — a standard and a low-carbon line, a base model and an efficient one. Then:

  implied cost per tonne  =  price difference / (carbon difference in tonnes)

Worked, from the chapter: $1.75 / (2.44/1000) = $717/t CO₂e, against internal carbon prices in corporate use of $50, $100, $190 and $250 a tonne, none of which clears it.

Now do the uncomfortable half. Find one behavioural or operational change available to you that removes carbon at no capital cost, estimate it honestly as a range, and put the two on the same page. The chapter's version: a 30 percent maintenance reduction removes 1.813 kg CO₂e/m² — 0.74× the whole gate-stage cut — for nothing.

Exercise 2.3 — The counterfactual spread (45 minutes)

Take any "we avoided X tonnes" claim you can find and recompute it against three different baselines. Use the IPCC AR5 lifecycle medians where the claim is about electricity: coal 820, gas combined cycle 490, EU-27 grid average 242, offshore wind 12, onshore wind 11 g CO₂e/kWh — and note the published ranges, which are wide: offshore wind 8.0–35, onshore 7.0–56, solar photovoltaic utility-scale 48 with a range of 18–180, hydropower 24 with a range of 1–2200.

That last range is the one to remember. A median of 24 with a maximum of 2200 is not a number, it is a distribution, and anybody quoting the median alone has made a choice.

---

PART THREE — DREAM AND DESIGN

Weeks 9–12: build the instrument you will keep

Exercise 3.1 — Your declaration card (2 hours)

Build the one-page card from the chapter as a template you will reuse for years. Four fields, nothing else: boundary, baseline, counterfactual, unit. Then fill it in three times:

• once for a product you already own and like;
• once for a claim you have personally made about your own behaviour;
• once for your institution's environmental pledge.

The third is the interesting one. University and employer net-zero pledges are almost always missing the counterfactual field and frequently missing the unit. Fill in what is there, mark what is not, and keep the card.

Exercise 3.2 — The restatement rule (one hour)

Write, in your own words, a rule for yourself: when I change a boundary, I say so and I restate the prior figure. Then apply it retroactively to one claim you have made — about your emissions, your spending, your time — where you quietly changed what counted. Restate it properly, both figures side by side.

This is the single habit that separates people whose numbers can be trusted from people whose numbers improve every year.

Exercise 3.3 — Read one declaration to somebody else (45 minutes)

Take the declaration you know best and walk one other person through it in ten minutes: the functional unit, the service life, the module table, the one line that complicates the headline. Watch where they get lost. Rewrite your explanation. Do it again with somebody else.

You will find the second telling is twice as good, and the third is the one you will use for the rest of your life.

---

PART FOUR — DESTINY AND DELIGHT

Weeks 13–15: make it hold, and enjoy it

Exercise 4.1 — The standing question (ongoing)

Adopt one question and ask it automatically, the way you check a price: what is the service life, and is the use-stage number annualised across it? It takes four seconds and it is the highest-yield question in this entire volume.

Exercise 4.2 — The generous reading (45 minutes)

Find a manufacturer who published something that made their own claim harder — a wide uncertainty range, an unflattering module, a stated assumption they could have omitted. Write two hundred words on what it cost them to publish it and what it bought them.

This exercise exists because the discipline in this chapter curdles very easily into cynicism, and cynicism is useless. The practice is only sustainable if you can also recognise good work when you are looking straight at it.

Exercise 4.3 — The pleasure of the seventh page

Notice, once, the specific feeling of finding on page seven of a document the table that complicates everything on page one — and of realising that somebody put it there on purpose, years ago, at expense, so that a stranger could disagree with them properly. Write one paragraph about it while it is fresh. That feeling is what keeps this practice alive after the term ends.

---

THE TERM PROJECT

One object, followed all the way

The brief. Choose one manufactured object. Produce a twelve- to fifteen-page study of it containing:

1. The object and its function. Photograph, description, and the functional unit you have chosen with your reason for choosing it.
2. The declaration, read. Every declared module, tabulated, with the undeclared ones named.
3. The six boundaries. Your own version of the chapter's table, computed, with the spread and any sign change.
4. The baseline. What the maker compares it with, sourced and dated — or, if they do not, what you would use and why.
5. The counterfactual. If the claim depends on one, state it as a scenario and compute the claim against three alternatives.
6. The implied abatement cost. With the comparison against a no-capital operational change.
7. The honest negative. One thing this object does worse than its alternative, or one condition its advantage depends on that is usually absent.
8. The recommendation. One page, addressed to a buyer, saying what you would actually do.

What makes it good. Not the verdict. The reproducibility. Every number should carry its unit and its source, and a reader with your study and the manufacturer's PDF should be able to rebuild every figure. If you have rounded, say so. If you have assumed, label it a scenario, as the chapter labels its thirty percent.

What makes it excellent. A finding the manufacturer would want. Not a gotcha — an observation that the arithmetic makes available and that the maker could act on. In this chapter's worked case, that finding is that the cleaning contract matters more than the yarn. Yours will be somewhere equally unglamorous.

---

SELF-ASSESSMENT

Score each honestly, one to five. Do it in week 1 and again in week 15 and keep both.

The last row is the one that matters most and the one that improves least by itself. A missing answer and a zero are not the same fact.

---

CARRYING IT FORWARD

Three things worth keeping past the end of the term.

The four-field card. It works on anything — a job offer, a business case, a medical claim, a policy. Boundary, baseline, counterfactual, unit. Most bad arguments you will meet for the rest of your life are missing one of them, and you will know which within a minute.

The annualisation reflex. A rate is not a total. Find the multiplier before you form a view. It is the same move whether the rate is kilograms per year, pounds per month or hours per week.

The generous first assumption. Start from the belief that the person who wrote the document was trying to tell you something true, and read until you are proven wrong. You will be proven wrong occasionally. You will be surprised far more often, and you will be a much better reader than someone who started suspicious.

---

APPRECIATIVE QUESTIONS FOR YOUR SEMINAR

1. Which declaration did the group find most generous, and what specifically made it so?
2. When has someone in this room changed their mind because of a number they computed themselves rather than one they were shown?
3. What would our institution's pledge look like if it carried all four declarations — and which of them would get stronger in the rewriting?
4. Where has a wide uncertainty range made a claim more credible to you rather than less, and why?

---

A NOTE ON READING RANGES

The single most misread thing in this whole literature is an uncertainty range, and it is worth ten minutes of deliberate practice because it will change how you read everything else.

The IPCC's Fifth Assessment gives lifecycle intensities as a median with a minimum and a maximum. Hydropower's median is 24 g CO₂e/kWh and its range is 1 to 2200. Solar photovoltaic at utility scale is 48, range 18 to 180. Offshore wind is 12, range 8.0 to 35; onshore wind is 11, range 7.0 to 56; nuclear is 12, range 3.7 to 110; dedicated biomass is 230, range 130 to 420; combined-cycle gas is 490, range 410 to 650; pulverised coal is 820, range 740 to 910.

Read that column of ranges once more and notice what it is telling you. Coal's range is narrow — a coal station is a coal station, wherever it is. Hydropower's range spans three orders of magnitude, because a reservoir in a cold rock valley and a reservoir flooding warm vegetation are physically different machines that happen to share a name.

A median without its range is a technology's average behaviour standing in for a specific project's actual behaviour. When somebody quotes you 24 for a particular dam, the honest question is not whether they are right. It is which end of a range from 1 to 2200 this dam sits at, and what determines that.

Exercise. Take three claims you collected in Exercise 1.3 and find, for each, whether the source publishes a range. Where it does, write both the point and the range. Where it does not, write range not published — and resist, firmly, the temptation to substitute a guess. A missing answer and a number are not the same fact, and a workbook that trains you to tell them apart has done the most useful thing it can do.

ONE LAST HABIT

At the end of every study you produce, write two short lists.

What I computed. Every figure you derived yourself, with its inputs.

What I could not check. Every figure you took on trust, and from whom.

The second list is the one that makes the first list believable. It is also, reliably, the list that gives you next term's project — because the thing you could not check this time is usually the thing worth learning to check next.